StepFun: Step 3.5 Flash
Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....
stepfun/step-3.5-flash
Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....
Specifications & capabilities
- Input
- text
- Output
- text
- Context window · tokens
- 262144
- Published model ID
- step-3.5-flash
- Capabilities
- promptCaching, reasoning, streaming, structuredOutputs, toolCalling
Evaluations
41–60 / 95
1,388.96
- source id
- lmarena
- leaderboard name
- LMArena Text Polish
- dimension
- general
- metric
- Arena rating
- rank
- 139
- sample size
- 1364
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1372.8259981678934
- confidence upper
- 1405.0947077276032
Original result ↗1,351.272
- source id
- lmarena
- leaderboard name
- LMArena Text Style Control Japanese
- dimension
- style
- metric
- Arena rating
- rank
- 139
- sample size
- 641
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1326.6381641732655
- confidence upper
- 1375.9058408217747
Original result ↗1,393.102
- source id
- lmarena
- leaderboard name
- LMArena Text Style Control German
- dimension
- style
- metric
- Arena rating
- rank
- 140
- sample size
- 1069
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1374.2030516731945
- confidence upper
- 1412.000065694564
Original result ↗1,419.895
- source id
- lmarena
- leaderboard name
- LMArena Text Style Control Industry Mathematical
- dimension
- style
- metric
- Arena rating
- rank
- 141
- sample size
- 3116
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1408.9627491199983
- confidence upper
- 1430.8274121475201
Original result ↗1,376.406
- source id
- lmarena
- leaderboard name
- LMArena Text Factuality Industry Entertainment And Sports And Media
- dimension
- factuality
- metric
- Arena rating
- rank
- 141
- sample size
- 9916
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1370.4744363585999
- confidence upper
- 1382.3367219687536
Original result ↗1,350.171
- source id
- lmarena
- leaderboard name
- LMArena Text Korean
- dimension
- general
- metric
- Arena rating
- rank
- 142
- sample size
- 1073
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1331.3681898303787
- confidence upper
- 1368.9733321720018
Original result ↗1,406.541
- source id
- lmarena
- leaderboard name
- LMArena Text Factuality Exclude Ties
- dimension
- factuality
- metric
- Arena rating
- rank
- 144
- sample size
- 43273
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1402.6201635189695
- confidence upper
- 1410.4609180053762
Original result ↗1,403.837
- source id
- lmarena
- leaderboard name
- LMArena Text Style Control Spanish
- dimension
- style
- metric
- Arena rating
- rank
- 145
- sample size
- 2005
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1389.8455091206247
- confidence upper
- 1417.8291824957446
Original result ↗1,420.228
- source id
- lmarena
- leaderboard name
- LMArena Text Hard Prompts English
- dimension
- general
- metric
- Arena rating
- rank
- 146
- sample size
- 16111
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1414.6407159707992
- confidence upper
- 1425.8142868560833
Original result ↗1,415.952
- source id
- lmarena
- leaderboard name
- LMArena Text Factuality Overall
- dimension
- factuality
- metric
- Arena rating
- rank
- 146
- sample size
- 59430
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1412.93730974349
- confidence upper
- 1418.9667041193588
Original result ↗1,437.43
- source id
- lmarena
- leaderboard name
- LMArena Text Factuality Hard Prompts
- dimension
- factuality
- metric
- Arena rating
- rank
- 147
- sample size
- 36890
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1433.6707280717378
- confidence upper
- 1441.1895636106135
Original result ↗1,416.881
- source id
- lmarena
- leaderboard name
- LMArena Text English
- dimension
- general
- metric
- Arena rating
- rank
- 147
- sample size
- 25049
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1412.1472203714775
- confidence upper
- 1421.614050853291
Original result ↗1,402.049
- source id
- lmarena
- leaderboard name
- LMArena Text Longer Query
- dimension
- general
- metric
- Arena rating
- rank
- 149
- sample size
- 22531
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1396.920139087082
- confidence upper
- 1407.1771080534454
Original result ↗1,408.829
- source id
- lmarena
- leaderboard name
- LMArena Text Industry Medicine And Healthcare
- dimension
- general
- metric
- Arena rating
- rank
- 150
- sample size
- 4184
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1399.2642075594097
- confidence upper
- 1418.3934549749224
Original result ↗1,407.783
- source id
- lmarena
- leaderboard name
- LMArena Text Industry Legal And Government
- dimension
- general
- metric
- Arena rating
- rank
- 150
- sample size
- 4612
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1398.6952773547437
- confidence upper
- 1416.8707176395096
Original result ↗1,404.434
- source id
- lmarena
- leaderboard name
- LMArena Text Multi Turn
- dimension
- general
- metric
- Arena rating
- rank
- 150
- sample size
- 10329
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1397.9656452171994
- confidence upper
- 1410.9016379553664
Original result ↗1,397.462
- source id
- lmarena
- leaderboard name
- LMArena Text Factuality Non English
- dimension
- factuality
- metric
- Arena rating
- rank
- 151
- sample size
- 34431
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1393.7028559301507
- confidence upper
- 1401.2220516552495
Original result ↗1,428.332
- source id
- lmarena
- leaderboard name
- LMArena Text Style Control Expert
- dimension
- style
- metric
- Arena rating
- rank
- 153
- sample size
- 5247
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1419.8232816304912
- confidence upper
- 1436.8401161202003
Original result ↗1,410.586
- source id
- lmarena
- leaderboard name
- LMArena Text Hard Prompts
- dimension
- general
- metric
- Arena rating
- rank
- 153
- sample size
- 37189
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1406.3246414836415
- confidence upper
- 1414.847327337032
Original result ↗1,406.925
- source id
- lmarena
- leaderboard name
- LMArena Text Math
- dimension
- general
- metric
- Arena rating
- rank
- 153
- sample size
- 3403
- evaluated at
- 2026-11-09T00:00:00.000Z
- source model id
- Step-3.5-Flash
- confidence lower
- 1396.713777277292
- confidence upper
- 1417.135952503252
Original result ↗Published prices
1–7 / 7
$0.09
- Inference provider
- deepInfra
- Platform
- DeepInfra
- Unit
- input_tokens_per_million
- Currency
- USD
$0.30
- Inference provider
- deepInfra
- Platform
- DeepInfra
- Unit
- output_tokens_per_million
- Currency
- USD
$0.10
- Inference provider
- siliconFlow
- Platform
- SiliconFlow
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing label
- Input tokens (per 1M)
$0.30
- Inference provider
- siliconFlow
- Platform
- SiliconFlow
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing label
- Output tokens (per 1M)
$0.02
- Inference provider
- stepFun
- Platform
- StepFun
- Unit
- cache_read_tokens_per_million
- Currency
- USD
- Billing label
- Cached input / 1M tokens
$0.10
- Inference provider
- stepFun
- Platform
- StepFun
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing label
- Input / 1M tokens (cache miss)
$0.30
- Inference provider
- stepFun
- Platform
- StepFun
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing label
- Output / 1M tokens
Provider specifications & access
1–2 / 2
- source model id
- step-3.5-flash
- hosting
- stepfun-global-hosted
- context length
- 256000
- capabilities
- promptCaching
- reasoning
- streaming
- structuredOutputs
- toolCalling
- source model id
- stepfun-ai/Step-3.5-Flash
- hosting
- siliconflow-hosted
- max output tokens
- 65536
Source specification observations
1–3 / 3
- field
- context_length
- value
- 262K
- scope
- source-model
- source id
- siliconFlow
- confidence
- 1
- field
- context_length
- value
- 256000
- normalized value
- 256000
- scope
- source-model
- source id
- stepFun
- confidence
- 1
- field
- max_output_tokens
- value
- 66K
- scope
- source-model
- source id
- siliconFlow
- confidence
- 1
Explore more models